Training in progress, step 8000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 966995080
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac71f6104740cd4c6f5b2f0a4f1d312da1f27f2b28f81745260c3b9565e2c13f
|
| 3 |
size 966995080
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1925064044
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa70ddee520870b9de82d68cf42fd025e0d52691a3ae84d400a9cf6b5eefc95d
|
| 3 |
size 1925064044
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bcd75decc8ec809bdd000c1a023eecd569d9a9775fe640822926fa2ab60021b
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d64c6a32c80af160202850f4c0903e8541ea45c1c0ea5e8f7b7cbe62296d013
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 16.
|
| 3 |
-
"best_model_checkpoint": "./whisper-small-taiwanese-hanzi/checkpoint-
|
| 4 |
-
"epoch": 2
|
| 5 |
"eval_steps": 1000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -2030,6 +2030,295 @@
|
|
| 2030 |
"eval_samples_per_second": 2.241,
|
| 2031 |
"eval_steps_per_second": 0.28,
|
| 2032 |
"step": 7000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2033 |
}
|
| 2034 |
],
|
| 2035 |
"logging_steps": 25,
|
|
@@ -2037,7 +2326,7 @@
|
|
| 2037 |
"num_input_tokens_seen": 0,
|
| 2038 |
"num_train_epochs": 4,
|
| 2039 |
"save_steps": 1000,
|
| 2040 |
-
"total_flos": 3.
|
| 2041 |
"train_batch_size": 8,
|
| 2042 |
"trial_name": null,
|
| 2043 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 16.252549215465933,
|
| 3 |
+
"best_model_checkpoint": "./whisper-small-taiwanese-hanzi/checkpoint-8000",
|
| 4 |
+
"epoch": 3.2,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 8000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 2030 |
"eval_samples_per_second": 2.241,
|
| 2031 |
"eval_steps_per_second": 0.28,
|
| 2032 |
"step": 7000
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 2.81,
|
| 2036 |
+
"grad_norm": 4.931020736694336,
|
| 2037 |
+
"learning_rate": 1.3541666666666667e-06,
|
| 2038 |
+
"loss": 0.1069,
|
| 2039 |
+
"step": 7025
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 2.82,
|
| 2043 |
+
"grad_norm": 4.510303497314453,
|
| 2044 |
+
"learning_rate": 1.3194444444444446e-06,
|
| 2045 |
+
"loss": 0.1088,
|
| 2046 |
+
"step": 7050
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 2.83,
|
| 2050 |
+
"grad_norm": 6.164842128753662,
|
| 2051 |
+
"learning_rate": 1.2847222222222222e-06,
|
| 2052 |
+
"loss": 0.1168,
|
| 2053 |
+
"step": 7075
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 2.84,
|
| 2057 |
+
"grad_norm": 4.769392490386963,
|
| 2058 |
+
"learning_rate": 1.25e-06,
|
| 2059 |
+
"loss": 0.1235,
|
| 2060 |
+
"step": 7100
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 2.85,
|
| 2064 |
+
"grad_norm": 5.1608710289001465,
|
| 2065 |
+
"learning_rate": 1.2152777777777778e-06,
|
| 2066 |
+
"loss": 0.1068,
|
| 2067 |
+
"step": 7125
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 2.86,
|
| 2071 |
+
"grad_norm": 6.107107639312744,
|
| 2072 |
+
"learning_rate": 1.1805555555555556e-06,
|
| 2073 |
+
"loss": 0.1213,
|
| 2074 |
+
"step": 7150
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 2.87,
|
| 2078 |
+
"grad_norm": 5.863134860992432,
|
| 2079 |
+
"learning_rate": 1.1458333333333333e-06,
|
| 2080 |
+
"loss": 0.1087,
|
| 2081 |
+
"step": 7175
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 2.88,
|
| 2085 |
+
"grad_norm": 4.510376453399658,
|
| 2086 |
+
"learning_rate": 1.111111111111111e-06,
|
| 2087 |
+
"loss": 0.1246,
|
| 2088 |
+
"step": 7200
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 2.89,
|
| 2092 |
+
"grad_norm": 6.96931791305542,
|
| 2093 |
+
"learning_rate": 1.076388888888889e-06,
|
| 2094 |
+
"loss": 0.1113,
|
| 2095 |
+
"step": 7225
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 2.9,
|
| 2099 |
+
"grad_norm": 6.812278747558594,
|
| 2100 |
+
"learning_rate": 1.0416666666666667e-06,
|
| 2101 |
+
"loss": 0.1392,
|
| 2102 |
+
"step": 7250
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 2.91,
|
| 2106 |
+
"grad_norm": 5.756344318389893,
|
| 2107 |
+
"learning_rate": 1.0069444444444447e-06,
|
| 2108 |
+
"loss": 0.0846,
|
| 2109 |
+
"step": 7275
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 2.92,
|
| 2113 |
+
"grad_norm": 3.710134506225586,
|
| 2114 |
+
"learning_rate": 9.722222222222224e-07,
|
| 2115 |
+
"loss": 0.101,
|
| 2116 |
+
"step": 7300
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 2.93,
|
| 2120 |
+
"grad_norm": 6.572783946990967,
|
| 2121 |
+
"learning_rate": 9.375000000000001e-07,
|
| 2122 |
+
"loss": 0.1141,
|
| 2123 |
+
"step": 7325
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 2.94,
|
| 2127 |
+
"grad_norm": 8.185981750488281,
|
| 2128 |
+
"learning_rate": 9.027777777777779e-07,
|
| 2129 |
+
"loss": 0.1207,
|
| 2130 |
+
"step": 7350
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 2.95,
|
| 2134 |
+
"grad_norm": 4.434142589569092,
|
| 2135 |
+
"learning_rate": 8.680555555555556e-07,
|
| 2136 |
+
"loss": 0.1206,
|
| 2137 |
+
"step": 7375
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 2.96,
|
| 2141 |
+
"grad_norm": 4.396365165710449,
|
| 2142 |
+
"learning_rate": 8.333333333333333e-07,
|
| 2143 |
+
"loss": 0.1235,
|
| 2144 |
+
"step": 7400
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 2.9699999999999998,
|
| 2148 |
+
"grad_norm": 8.427244186401367,
|
| 2149 |
+
"learning_rate": 7.986111111111111e-07,
|
| 2150 |
+
"loss": 0.1131,
|
| 2151 |
+
"step": 7425
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 2.98,
|
| 2155 |
+
"grad_norm": 5.311284065246582,
|
| 2156 |
+
"learning_rate": 7.63888888888889e-07,
|
| 2157 |
+
"loss": 0.1141,
|
| 2158 |
+
"step": 7450
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 2.99,
|
| 2162 |
+
"grad_norm": 3.4007761478424072,
|
| 2163 |
+
"learning_rate": 7.291666666666667e-07,
|
| 2164 |
+
"loss": 0.1031,
|
| 2165 |
+
"step": 7475
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 3.0,
|
| 2169 |
+
"grad_norm": 4.035045146942139,
|
| 2170 |
+
"learning_rate": 6.944444444444446e-07,
|
| 2171 |
+
"loss": 0.0908,
|
| 2172 |
+
"step": 7500
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 3.01,
|
| 2176 |
+
"grad_norm": 2.659799337387085,
|
| 2177 |
+
"learning_rate": 6.597222222222223e-07,
|
| 2178 |
+
"loss": 0.061,
|
| 2179 |
+
"step": 7525
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 3.02,
|
| 2183 |
+
"grad_norm": 3.8670284748077393,
|
| 2184 |
+
"learning_rate": 6.25e-07,
|
| 2185 |
+
"loss": 0.0687,
|
| 2186 |
+
"step": 7550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 3.03,
|
| 2190 |
+
"grad_norm": 3.3453280925750732,
|
| 2191 |
+
"learning_rate": 5.902777777777778e-07,
|
| 2192 |
+
"loss": 0.0619,
|
| 2193 |
+
"step": 7575
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 3.04,
|
| 2197 |
+
"grad_norm": 4.7413249015808105,
|
| 2198 |
+
"learning_rate": 5.555555555555555e-07,
|
| 2199 |
+
"loss": 0.0701,
|
| 2200 |
+
"step": 7600
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 3.05,
|
| 2204 |
+
"grad_norm": 3.685469627380371,
|
| 2205 |
+
"learning_rate": 5.208333333333334e-07,
|
| 2206 |
+
"loss": 0.0712,
|
| 2207 |
+
"step": 7625
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 3.06,
|
| 2211 |
+
"grad_norm": 1.9484667778015137,
|
| 2212 |
+
"learning_rate": 4.861111111111112e-07,
|
| 2213 |
+
"loss": 0.0699,
|
| 2214 |
+
"step": 7650
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 3.07,
|
| 2218 |
+
"grad_norm": 4.542367458343506,
|
| 2219 |
+
"learning_rate": 4.5138888888888893e-07,
|
| 2220 |
+
"loss": 0.0608,
|
| 2221 |
+
"step": 7675
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 3.08,
|
| 2225 |
+
"grad_norm": 3.3884780406951904,
|
| 2226 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 2227 |
+
"loss": 0.0629,
|
| 2228 |
+
"step": 7700
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 3.09,
|
| 2232 |
+
"grad_norm": 3.8800745010375977,
|
| 2233 |
+
"learning_rate": 3.819444444444445e-07,
|
| 2234 |
+
"loss": 0.0627,
|
| 2235 |
+
"step": 7725
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 3.1,
|
| 2239 |
+
"grad_norm": 3.6111207008361816,
|
| 2240 |
+
"learning_rate": 3.472222222222223e-07,
|
| 2241 |
+
"loss": 0.0616,
|
| 2242 |
+
"step": 7750
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 3.11,
|
| 2246 |
+
"grad_norm": 4.225738525390625,
|
| 2247 |
+
"learning_rate": 3.125e-07,
|
| 2248 |
+
"loss": 0.0701,
|
| 2249 |
+
"step": 7775
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 3.12,
|
| 2253 |
+
"grad_norm": 4.790248394012451,
|
| 2254 |
+
"learning_rate": 2.7777777777777776e-07,
|
| 2255 |
+
"loss": 0.0761,
|
| 2256 |
+
"step": 7800
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 3.13,
|
| 2260 |
+
"grad_norm": 5.802876949310303,
|
| 2261 |
+
"learning_rate": 2.430555555555556e-07,
|
| 2262 |
+
"loss": 0.0647,
|
| 2263 |
+
"step": 7825
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 3.14,
|
| 2267 |
+
"grad_norm": 4.658420085906982,
|
| 2268 |
+
"learning_rate": 2.0833333333333333e-07,
|
| 2269 |
+
"loss": 0.0605,
|
| 2270 |
+
"step": 7850
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 3.15,
|
| 2274 |
+
"grad_norm": 3.0659847259521484,
|
| 2275 |
+
"learning_rate": 1.7361111111111115e-07,
|
| 2276 |
+
"loss": 0.062,
|
| 2277 |
+
"step": 7875
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 3.16,
|
| 2281 |
+
"grad_norm": 2.227428436279297,
|
| 2282 |
+
"learning_rate": 1.3888888888888888e-07,
|
| 2283 |
+
"loss": 0.0572,
|
| 2284 |
+
"step": 7900
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 3.17,
|
| 2288 |
+
"grad_norm": 2.0126748085021973,
|
| 2289 |
+
"learning_rate": 1.0416666666666667e-07,
|
| 2290 |
+
"loss": 0.056,
|
| 2291 |
+
"step": 7925
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 3.18,
|
| 2295 |
+
"grad_norm": 5.308594703674316,
|
| 2296 |
+
"learning_rate": 6.944444444444444e-08,
|
| 2297 |
+
"loss": 0.0728,
|
| 2298 |
+
"step": 7950
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 3.19,
|
| 2302 |
+
"grad_norm": 2.8052618503570557,
|
| 2303 |
+
"learning_rate": 3.472222222222222e-08,
|
| 2304 |
+
"loss": 0.0744,
|
| 2305 |
+
"step": 7975
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 3.2,
|
| 2309 |
+
"grad_norm": 3.2474498748779297,
|
| 2310 |
+
"learning_rate": 0.0,
|
| 2311 |
+
"loss": 0.0678,
|
| 2312 |
+
"step": 8000
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 3.2,
|
| 2316 |
+
"eval_cer": 16.252549215465933,
|
| 2317 |
+
"eval_loss": 0.2705218493938446,
|
| 2318 |
+
"eval_runtime": 1747.8851,
|
| 2319 |
+
"eval_samples_per_second": 2.252,
|
| 2320 |
+
"eval_steps_per_second": 0.281,
|
| 2321 |
+
"step": 8000
|
| 2322 |
}
|
| 2323 |
],
|
| 2324 |
"logging_steps": 25,
|
|
|
|
| 2326 |
"num_input_tokens_seen": 0,
|
| 2327 |
"num_train_epochs": 4,
|
| 2328 |
"save_steps": 1000,
|
| 2329 |
+
"total_flos": 3.693893124096e+19,
|
| 2330 |
"train_batch_size": 8,
|
| 2331 |
"trial_name": null,
|
| 2332 |
"trial_params": null
|