learning curves for OC patent.Rmd

```{r}
# 📦 Load libraries
library(ggplot2)
library(dplyr)
library(readr)
library(lubridate)
library(stringr)




# 📁 Define CSV paths for each week
csv_paths <- list(
  "Week 6" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3.5hrs_day1_24.2.25/decoded_datalog_22425_3543.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day2_25.2.25/decoded_datalog_22525_3545.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day3_26.2.25/decoded_datalog_22625_0815.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day4_27.2.25/decoded_datalog_22725_01220.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day5_28.2.25/decoded_datalog_22825_12430.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_4.5hrs_day6_1.3.25/decoded_datalog_3125_1449.csv"
  ),
  "Week 7" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day1_3.3.25/decoded_datalog_3325_01522.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day2_4.3.25/decoded_datalog_3425_05739.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day3_5.3.25/decoded_datalog_3525_52435.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day4_7.3.25/decoded_datalog_3725_22737.csv"
  ),
  "Week 9" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_3hrs_day1_18.3.25/decoded_datalog_31825_11551.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_2hrs_2sessions_day2_19.3.25/session 1_2.5hrs/decoded_datalog_31925_13231.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_2hrs_2sessions_day2_19.3.25/session 2_1hr/decoded_datalog_31925_172212.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_20.3.25/session 1/decoded_datalog_32025_131813.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_20.3.25/session 2/decoded_datalog_32025_162149.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 2 (Unmarked)/Training CSVs/week9_1hr_2session_day4_21.3.25/session 1/decoded_datalog_32125_11143.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_21.3.25/session 2/decoded_datalog_32125_17924.csv"
  ),
  "Week 10" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day2_25.3.25/decoded_datalog_32525_131213.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2.5hrs_day3_26.3.25/decoded_datalog_32625_133243.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_1.5hrs_day4_27.3.25/decoded_datalog_32725_154415.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day5_28.3.25/decoded_datalog_32825_111358.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day5_29.3.25/decoded_datalog_32925_13831.csv"
  ),
  "Week 11" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_1.5hrs_31.3.25/decoded_datalog_33125_154911.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_2hrs_1.4.25/decoded_datalog_4125_12156.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_2.5hrs_2.4.25/decoded_datalog_4225_132231.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_2hrs_day4_M_3.4.25/decoded_datalog_4325_111427.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_day5_4.4.25/decoded_datalog_4425_133254.csv"
  )
)

# 🔁 Function to compute RM per NP from each session's first 1 hour
compute_learning_metric <- function(file, week) {
  df <- read_csv(file, show_col_types = FALSE)
  df$Timestamp <- as.POSIXct(paste(df$Y, df$M, df$D, df$HR, df$MIN, df$SEC),
                              format = "%Y %m %d %H %M %S")
  start_time <- min(df$Timestamp, na.rm = TRUE)
  df <- df %>% filter(Timestamp <= start_time + 3600)

  df <- df %>% mutate(
    NP = ifelse(Decoded_NosePoke == "NP", 1, 0),
    RM = ifelse(Decoded_Reward == "RM", 1, 0)
  )

  rm_count <- sum(df$RM, na.rm = TRUE)
  np_count <- sum(df$NP, na.rm = TRUE)

  tibble(
    Week = week,
    RM = rm_count,
    NP = np_count,
    RM_per_NP = ifelse(np_count == 0, NA, rm_count / np_count)
  )
}

# 🧮 Apply across all weeks and files
library(purrr)
learning_data <- map2_dfr(names(csv_paths), csv_paths, function(week, paths) {
  map_dfr(paths, compute_learning_metric, week = week)
})

# 📊 Summarize
summary_df <- learning_data %>%
  group_by(Week) %>%
  summarise(
    Mean = mean(RM_per_NP, na.rm = TRUE),
    SD = sd(RM_per_NP, na.rm = TRUE),
    N = n()
  )

# 📈 Plot learning curve with error bars
ggplot(summary_df, aes(x = Week, y = Mean, group = 1)) +
  geom_point(size = 4, color = "#0073C2FF") +
  geom_line(color = "#0073C2FF", size = 1.5) +
  geom_errorbar(aes(ymin = Mean - SD, ymax = Mean + SD), width = 0.2, color = "darkgray") +
  labs(
    title = "Learning Curve: Harvests per Nose Poke (RM/NP)",
    x = "Training Week",
    y = "Mean RM per NP ± SD"
  ) +
  theme_minimal() +
  theme(
    axis.text = element_text(size = 12),
    axis.title = element_text(size = 14),
    plot.title = element_text(size = 16, face = "bold")
  )

```
```{r}
# 📦 Load libraries
library(ggplot2)
library(dplyr)
library(readr)
library(lubridate)
library(stringr)
library(purrr)

# 📁 Define CSV paths grouped by week
csv_paths <- list(
  "Week 6" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3.5hrs_day1_24.2.25/decoded_datalog_22425_3543.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day2_25.2.25/decoded_datalog_22525_3545.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day3_26.2.25/decoded_datalog_22625_0815.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day4_27.2.25/decoded_datalog_22725_01220.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day5_28.2.25/decoded_datalog_22825_12430.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_4.5hrs_day6_1.3.25/decoded_datalog_3125_1449.csv"
  ),
  "Week 7" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day1_3.3.25/decoded_datalog_3325_01522.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day2_4.3.25/decoded_datalog_3425_05739.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day3_5.3.25/decoded_datalog_3525_52435.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day4_7.3.25/decoded_datalog_3725_22737.csv"
  ),
  "Week 9" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_3hrs_day1_18.3.25/decoded_datalog_31825_11551.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_2hrs_2sessions_day2_19.3.25/session 1_2.5hrs/decoded_datalog_31925_13231.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_2hrs_2sessions_day2_19.3.25/session 2_1hr/decoded_datalog_31925_172212.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_20.3.25/session 1/decoded_datalog_32025_131813.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_20.3.25/session 2/decoded_datalog_32025_162149.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 2 (Unmarked)/Training CSVs/week9_1hr_2session_day4_21.3.25/session 1/decoded_datalog_32125_11143.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_21.3.25/session 2/decoded_datalog_32125_17924.csv"
  ),
  "Week 10" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day2_25.3.25/decoded_datalog_32525_131213.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2.5hrs_day3_26.3.25/decoded_datalog_32625_133243.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_1.5hrs_day4_27.3.25/decoded_datalog_32725_154415.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day5_28.3.25/decoded_datalog_32825_111358.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day5_29.3.25/decoded_datalog_32925_13831.csv"
  ),
  "Week 11" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_1.5hrs_31.3.25/decoded_datalog_33125_154911.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_2hrs_1.4.25/decoded_datalog_4125_12156.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_2.5hrs_2.4.25/decoded_datalog_4225_132231.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_2hrs_day4_M_3.4.25/decoded_datalog_4325_111427.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week11_day5_4.4.25/decoded_datalog_4425_133254.csv"
  )
)

# 🔁 Function to compute RM/NP from first 1 hour of data
compute_learning_metric <- function(file, week) {
  df <- read_csv(file, show_col_types = FALSE)
  df$Timestamp <- as.POSIXct(paste(df$Y, df$M, df$D, df$HR, df$MIN, df$SEC),
                             format = "%Y %m %d %H %M %S")
  start_time <- min(df$Timestamp, na.rm = TRUE)
  df <- df %>% filter(Timestamp <= start_time + 3600)

  df <- df %>%
    mutate(
      NP = ifelse(Decoded_NosePoke == "NP", 1, 0),
      RM = ifelse(Decoded_Reward == "RM", 1, 0)
    )

  rm_count <- sum(df$RM, na.rm = TRUE)
  np_count <- sum(df$NP, na.rm = TRUE)

  tibble(
    Week = week,
    RM = rm_count,
    NP = np_count,
    RM_per_NP = ifelse(np_count == 0, NA, rm_count / np_count)
  )
}

# 🧠 Apply across all weeks and files
learning_data <- map2_dfr(names(csv_paths), csv_paths, function(week, paths) {
  map_dfr(paths, compute_learning_metric, week = week)
})

# 📊 Summarize
summary_df <- learning_data %>%
  group_by(Week) %>%
  summarise(
    Mean = mean(RM_per_NP, na.rm = TRUE),
    SD = sd(RM_per_NP, na.rm = TRUE),
    N = n()
  )

# 📈 Plot learning curve
ggplot(summary_df, aes(x = Week, y = Mean, group = 1)) +
  geom_point(size = 4, color = "#009E73") +
  geom_line(color = "#009E73", size = 1.2) +
  geom_errorbar(aes(ymin = Mean - SD, ymax = Mean + SD), width = 0.2, color = "gray30") +
  labs(
    title = "Learning Curve: Harvests per Nose Poke (RM/NP)",
    x = "Training Week",
    y = "Mean RM per NP ± SD"
  ) +
  theme_minimal() +
  theme(
    axis.text = element_text(size = 12),
    axis.title = element_text(size = 14),
    plot.title = element_text(size = 16, face = "bold")
  )

```
```{r}
# 📦 Load required libraries
library(ggplot2)
library(dplyr)
library(readr)
library(purrr)
library(lubridate)
library(forcats)

# 📂 Define correct chronological order
week_order <- c("Week 6", "Week 7", "Week 9", "Week 10", "Week 11")

# 🧮 Use already computed learning_data from previous chunk
# If not, recompute using compute_learning_metric and map2_dfr like earlier

# 🧮 Summarise by week and reorder
summary_df <- learning_data %>%
  group_by(Week) %>%
  summarise(
    Mean = mean(RM_per_NP, na.rm = TRUE),
    SD = sd(RM_per_NP, na.rm = TRUE),
    N = n()
  ) %>%
  mutate(Week = factor(Week, levels = week_order)) %>%
  arrange(Week)

# 📈 Final R Plot with chronological weeks
ggplot(summary_df, aes(x = Week, y = Mean, group = 1)) +
  geom_point(size = 4, color = "#F8766D") +
  geom_line(color = "#F8766D", size = 1.5) +
  geom_errorbar(aes(ymin = Mean - SD, ymax = Mean + SD), width = 0.2, color = "gray50") +
  labs(
    title = "Learning Curve: Harvests per Nose Poke (RM/NP)",
    x = "Training Week (Chronological)",
    y = "Mean RM per NP ± SD"
  ) +
  theme_minimal() +
  theme(
    axis.text = element_text(size = 12),
    axis.title = element_text(size = 14),
    plot.title = element_text(size = 16, face = "bold")
  )

```

```{r}
# 📦 Load required libraries
library(ggplot2)
library(dplyr)

# 🧪 Create summary data frame
summary_df <- data.frame(
  Week = factor(c("Week 6", "Week 7", "Week 9", "Week 10", "Week 11"),
                levels = c("Week 6", "Week 7", "Week 9", "Week 10", "Week 11"),
                ordered = TRUE),
  Mean = c(2.7, 2.6, 8.2, 11.3, 8.3),
  SD = c(1.1, 1.7, 3.5, 3.2, 3.4)
)

# 📈 Plot learning curve
ggplot(summary_df, aes(x = Week, y = Mean, group = 1)) +
  geom_point(size = 4, color = "#20A387FF") +
  geom_line(size = 1.5, color = "#20A387FF") +
  geom_errorbar(aes(ymin = Mean - SD, ymax = Mean + SD), width = 0.2, color = "darkgray") +
  labs(
    title = "Learning Curve: Harvests per Nose Poke (RM/NP)",
    x = "Training Week",
    y = "Mean RM per NP ± SD"
  ) +
  theme_minimal() +
  theme(
    axis.text = element_text(size = 12),
    axis.title = element_text(size = 14),
    plot.title = element_text(size = 16, face = "bold")
  )

```

```{r}
# 📦 Load necessary libraries
library(ggplot2)
library(dplyr)
library(readr)
library(lubridate)
library(stringr)
library(purrr)

# 📁 Define CSV paths for each week
csv_paths <- list(
  "Week 6" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3.5hrs_day1_24.2.25/decoded_datalog_22425_3543.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day2_25.2.25/decoded_datalog_22525_3545.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day3_26.2.25/decoded_datalog_22625_0815.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day4_27.2.25/decoded_datalog_22725_01220.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day5_28.2.25/decoded_datalog_22825_12430.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_4.5hrs_day6_1.3.25/decoded_datalog_3125_1449.csv"
  ),
  "Week 7" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day1_3.3.25/decoded_datalog_3325_01522.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day2_4.3.25/decoded_datalog_3425_05739.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day3_5.3.25/decoded_datalog_3525_52435.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day4_7.3.25/decoded_datalog_3725_22737.csv"
  ),
  "Week 9" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_3hrs_day1_18.3.25/decoded_datalog_31825_11551.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_2hrs_2sessions_day2_19.3.25/session 1_2.5hrs/decoded_datalog_31925_13231.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_2hrs_2sessions_day2_19.3.25/session 2_1hr/decoded_datalog_31925_172212.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_20.3.25/session 1/decoded_datalog_32025_131813.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_20.3.25/session 2/decoded_datalog_32025_162149.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 2 (Unmarked)/Training CSVs/week9_1hr_2session_day4_21.3.25/session 1/decoded_datalog_32125_11143.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_21.3.25/session 2/decoded_datalog_32125_17924.csv"
  ),
  "Week 10" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day2_25.3.25/decoded_datalog_32525_131213.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2.5hrs_day3_26.3.25/decoded_datalog_32625_133243.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_1.5hrs_day4_27.3.25/decoded_datalog_32725_154415.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day5_28.3.25/decoded_datalog_32825_111358.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day5_29.3.25/decoded_datalog_32925_13831.csv"
  )
)

# 🔁 Compute RM/NP for first hour from each CSV
compute_learning_metric <- function(file, week) {
  df <- read_csv(file, show_col_types = FALSE)
  df$Timestamp <- as.POSIXct(paste(df$Y, df$M, df$D, df$HR, df$MIN, df$SEC), format = "%Y %m %d %H %M %S")
  start_time <- min(df$Timestamp, na.rm = TRUE)
  df <- df %>% filter(Timestamp <= start_time + 3600)  # First 1 hour
  
  df <- df %>% mutate(
    NP = ifelse(Decoded_NosePoke == "NP", 1, 0),
    RM = ifelse(Decoded_Reward == "RM", 1, 0)
  )

  rm_count <- sum(df$RM, na.rm = TRUE)
  np_count <- sum(df$NP, na.rm = TRUE)

  tibble(
    Week = week,
    RM = rm_count,
    NP = np_count,
    RM_per_NP = ifelse(np_count == 0, NA, rm_count / np_count)
  )
}

# 📊 Apply across all files
learning_data <- map2_dfr(names(csv_paths), csv_paths, function(week, paths) {
  map_dfr(paths, compute_learning_metric, week = week)
})

# 📈 Summarize per week
summary_df <- learning_data %>%
  mutate(Week = factor(Week, levels = c("Week 6", "Week 7", "Week 9", "Week 10", "Week 11"))) %>%
  group_by(Week) %>%
  summarise(
    Mean = mean(RM_per_NP, na.rm = TRUE),
    SD = sd(RM_per_NP, na.rm = TRUE),
    N = n()
  )

# 📉 Plot learning curve
ggplot(summary_df, aes(x = Week, y = Mean, group = 1)) +
  geom_point(size = 4, color = "#E69F00") +
  geom_line(color = "#E69F00", size = 1.5) +
  geom_errorbar(aes(ymin = Mean - SD, ymax = Mean + SD), width = 0.2, color = "gray40") +
  labs(
    title = "Learning Curve (RM per NP)",
    x = "Training Week",
    y = "Mean RM per NP ± SD"
  ) +
  theme_minimal() +
  theme(
    axis.text = element_text(size = 12),
    axis.title = element_text(size = 14),
    plot.title = element_text(size = 16, face = "bold")
  )
```
```{r}
# 📦 Load libraries
library(ggplot2)
library(dplyr)
library(readr)
library(lubridate)
library(purrr)
library(stringr)

# 🗂 Define all CSV paths as a named list
csv_paths <- list(
  "Week 6" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3.5hrs_day1_24.2.25/decoded_datalog_22425_3543.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day2_25.2.25/decoded_datalog_22525_3545.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day3_26.2.25/decoded_datalog_22625_0815.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day4_27.2.25/decoded_datalog_22725_01220.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_3hrs_day5_28.2.25/decoded_datalog_22825_12430.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week6_4.5hrs_day6_1.3.25/decoded_datalog_3125_1449.csv"
  ),
  "Week 7" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day1_3.3.25/decoded_datalog_3325_01522.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day2_4.3.25/decoded_datalog_3425_05739.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day3_5.3.25/decoded_datalog_3525_52435.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week7_3hrs_day4_7.3.25/decoded_datalog_3725_22737.csv"
  ),
  "Week 9" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_3hrs_day1_18.3.25/decoded_datalog_31825_11551.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_2hrs_2sessions_day2_19.3.25/session 1_2.5hrs/decoded_datalog_31925_13231.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_2hrs_2sessions_day2_19.3.25/session 2_1hr/decoded_datalog_31925_172212.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_20.3.25/session 1/decoded_datalog_32025_131813.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_20.3.25/session 2/decoded_datalog_32025_162149.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 2 (Unmarked)/Training CSVs/week9_1hr_2session_day4_21.3.25/session 1/decoded_datalog_32125_11143.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week9_1hr_2sessions_21.3.25/session 2/decoded_datalog_32125_17924.csv"
  ),
  "Week 10" = c(
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day2_25.3.25/decoded_datalog_32525_131213.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2.5hrs_day3_26.3.25/decoded_datalog_32625_133243.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_1.5hrs_day4_27.3.25/decoded_datalog_32725_154415.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day5_28.3.25/decoded_datalog_32825_111358.csv",
    "F:/Shrutika-Operant_Chamber-Training Data/Mouse 1 (Marked)/Training CSVs/week10_2hrs_day5_29.3.25/decoded_datalog_32925_13831.csv"
  )
)

# 🧮 Function to extract 1st hour event counts from a file
process_file <- function(file, week) {
  df <- tryCatch(read_csv(file, show_col_types = FALSE), error = function(e) return(NULL))
  if (is.null(df)) return(NULL)
  
  df$Timestamp <- as.POSIXct(paste(df$Y, df$M, df$D, df$HR, df$MIN, df$SEC), format = "%Y %m %d %H %M %S")
  start_time <- min(df$Timestamp, na.rm = TRUE)
  df <- df %>% filter(Timestamp <= start_time + hours(1))

  tibble(
    Week = week,
    NP = sum(df$Decoded_NosePoke == "NP", na.rm = TRUE),
    L1 = sum(df$Decoded_Lever_1 == "L1", na.rm = TRUE),
    L2 = sum(df$Decoded_Lever_2 == "L2", na.rm = TRUE),
    RM = sum(df$Decoded_Reward == "RM", na.rm = TRUE)
  )
}

# 🔁 Apply function to all CSVs and tag weeks
all_data <- map2_dfr(names(csv_paths), csv_paths, function(week, paths) {
  map_dfr(paths, process_file, week = week)
})

# 🧠 Convert Week into ordered factor for correct plotting order
all_data$Week <- factor(all_data$Week, levels = c("Week 6", "Week 7", "Week 9", "Week 10", "Week 11"))

# 📊 Function to plot with SD error bars
plot_metric <- function(metric_label) {
  summary_data <- all_data %>%
    group_by(Week) %>%
    summarise(
      Mean = mean(.data[[metric_label]], na.rm = TRUE),
      SD = sd(.data[[metric_label]], na.rm = TRUE)
    )
  
  ggplot(all_data, aes(x = Week, y = .data[[metric_label]])) +
    geom_point(position = position_jitter(width = 0.1), size = 3, color = "#0072B2", alpha = 0.6) +
    geom_line(data = summary_data, aes(x = Week, y = Mean, group = 1), color = "#D55E00", size = 1.2) +
    geom_point(data = summary_data, aes(x = Week, y = Mean), size = 4, color = "#D55E00") +
    geom_errorbar(data = summary_data, aes(x = Week, ymin = Mean - SD, ymax = Mean + SD), width = 0.2, color = "gray40") +
    labs(
      title = paste("Weekly", metric_label, "Summary"),
      x = "Training Week",
      y = paste("Total", metric_label, "(± SD)")
    ) +
    theme_minimal(base_size = 14)
}

# ⏱️ Summarise total NP per week with SD
summary_np <- master_data %>%
  group_by(Week) %>%
  summarise(
    Mean_NP = mean(NP, na.rm = TRUE),
    SD_NP = sd(NP, na.rm = TRUE),
    .groups = "drop"
  )

# ✅ Plotting with correct column names
ggplot(summary_np, aes(x = Week, y = Mean_NP, group = 1)) +
  geom_point(color = "steelblue", size = 3) +
  geom_line(color = "steelblue", size = 1) +
  geom_errorbar(aes(ymin = Mean_NP - SD_NP, ymax = Mean_NP + SD_NP), width = 0.2) +
  labs(
    title = "Total Nose Pokes Across Weeks",
    x = "Week",
    y = "Mean Nose Pokes ± SD"
  ) +
  theme_minimal()

```